On the Application of Stable Generalized Finite Element Method for Quasilinear Elliptic Two-Point BVP

Tilsa Aryeni, Quanling Deng, Victor Ginting

Abstract

In this paper, we discuss the application of the Generalized Finite Element Method (GFEM) to approximate the solutions of quasilinear elliptic equations with multiple interfaces in one dimensional space. The problem is characterized by spatial discontinuity of the elliptic coefficient that depends on the unknown solution. It is known that unless the partition of the domain matches the discontinuity configuration, accuracy of standard finite element techniques significantly deteriorates and standard refinement of the partition may not suffice. The GFEM is a viable alternative to overcome this predicament. It is based on the construction of certain enrichment functions supplied to the standard space that capture effects of the discontinuity. This approach is called stable (SGFEM) if it maintains an optimal rate of convergence and the conditioning of GFEM is not worse than that of the standard FEM. A convergence analysis is derived and performance of the method is illustrated by several numerical examples. Furthermore, it is known that typical global formulations such as FEMs do not enjoy the numerical local conservation property that is crucial in many conservation law-based applications. To remedy this issue, a Lagrange multiplier technique is adopted to enforce the local conservation. A numerical example is given to demonstrate the performance of proposed technique.

Introduction

Modeling fluid flow through heterogeneous media, such as layered soils in hydrology or petroleum engineering, often involves solving quasilinear elliptic equations. In these systems, the conductivity coefficient $\kappa(x, u)$ can change abruptly at material interfaces. Standard Finite Element Methods (FEM) struggle with such discontinuities unless the mesh is perfectly aligned with the interface. Misalignment leads to a significant loss of accuracy. While mesh refinement is a possible solution, it becomes computationally prohibitive for problems with moving interfaces or in stochastic simulations where interface locations vary.

Methodology: Stable Generalized Finite Element Method (SGFEM)

To address these challenges, the authors utilize the Stable Generalized Finite Element Method (SGFEM). The core idea is to "enrich" the standard polynomial basis functions with auxiliary functions that mimic the solution's behavior near discontinuities.

Enrichment and Stability

A significant hurdle in Generalized FEM is the potential for ill-conditioned matrices, where small numerical errors can lead to large instabilities. To ensure stability, the authors construct enrichment functions by subtracting the linear interpolant from the raw enrichment function. This modification ensures that the resulting SGFEM maintains the same conditioning as standard FEM while achieving optimal convergence.

Construction of the stable enrichment function $w_\tau$ (right) by subtracting the linear interpolant from the raw enrichment function $w_\tau^*$ (left).

The enriched basis functions, such as $\varphi_{k,E}$, are then formed by multiplying these stable enrichment functions with the standard partition of unity functions. This allows the method to capture the interface effects locally within the elements containing the discontinuity.

The resulting enriched basis function $\varphi_{k,E}$ used to capture interface discontinuities.

Local Conservation and Nonlinearity

Since the problem is quasilinear—meaning the conductivity coefficients depend on the unknown solution itself—the authors employ an iterative Newton's method to solve the resulting nonlinear algebraic system. Furthermore, standard FEM formulations often fail to satisfy local conservation—a physical requirement that mass balance is maintained over any control volume. The authors treat this conservation requirement as a constraint and employ a Lagrange multiplier technique to enforce it. This ensures that the numerical solution is physically consistent, which is vital for applications based on conservation laws.

Numerical Results

The method was evaluated using synthetic 1D benchmarks with multiple interfaces and varying conductivity models (e.g., $\kappa(x, u) = e^{au}$).

  • Optimal Convergence: Numerical tests confirmed theoretical predictions. For a polynomial degree $p$, the error decreases at a rate of $h^p$ in the $H^1$ semi-norm and $h^{p+1}$ in the $L^2$ norm, even when interfaces do not align with mesh nodes.
  • High Contrast Robustness: The SGFEM remained stable and accurate even with high contrast ratios, where material properties differed by a factor of over 2,600.
  • Conservation Performance: While standard SGFEM exhibited local conservation errors, the Lagrange multiplier approach reduced these errors to near-zero (machine precision, $\sim 10^{-12}$), effectively satisfying the physical mass balance.

Conclusion and Future Work

The SGFEM provides a robust, mesh-independent framework for solving quasilinear problems with interfaces. By incorporating stable enrichment and Lagrange multipliers, the authors achieve both numerical stability and physical consistency.

Future research will focus on extending this 1D proof-of-concept to higher dimensions (2D and 3D), which involves more complex enrichment function constructions. Additionally, the authors plan to apply this method to time-dependent problems, such as the Richards equation, to model dynamic fluid infiltration in porous media.

Summary updated

Abstract

In this paper, we discuss the application of the Generalized Finite Element Method (GFEM) to approximate the solutions of quasilinear elliptic equations with multiple interfaces in one dimensional space. The problem is characterized by spatial discontinuity of the elliptic coefficient that depends on the unknown solution. It is known that unless the partition of the domain matches the discontinuity configuration, accuracy of standard finite element techniques significantly deteriorates and standard refinement of the partition may not suffice. The GFEM is a viable alternative to overcome this predicament. It is based on the construction of certain enrichment functions supplied to the standard space that capture effects of the discontinuity. This approach is called stable (SGFEM) if it maintains an optimal rate of convergence and the conditioning of GFEM is not worse than that of the standard FEM. A convergence analysis is derived and performance of the method is illustrated by several numerical examples. Furthermore, it is known that typical global formulations such as FEMs do not enjoy the numerical local conservation property that is crucial in many conservation law-based applications. To remedy this issue, a Lagrange multiplier technique is adopted to enforce the local conservation. A numerical example is given to demonstrate the performance of proposed technique.
BibTeX
@article{aryeni2022application,
  author = {Tilsa Aryeni and Quanling Deng and Victor Ginting},
  title = {{On the Application of Stable Generalized Finite Element Method for Quasilinear Elliptic Two-Point BVP}},
  journal = {J. Sci. Comput.},
  year = {2022},
  url = {https://openreview.net/forum?id=oifkFHscYDsR},
}
Figure